ErbB-2-positive breast cancer is currently treated with therapeutic agents trastuzumab (Carter et al, 1992
) and lapatinib (Nahta et al, 2007
), or current preclinical studies (Rimawi et al, 2011
) and clinical trials are investigating the combination of trastuzumab plus lapatinib (Baselga et al, 2010
). Although trastuzumab plus chemotherapy has been successful in the treatment of ErbB-2-positive breast cancer, some patients will not respond to this drug and, among responders, 15% will have disease recurrence and ultimately progression (Cobleigh et al, 1999
; Vogel et al, 2002
). Thus, trastuzumab resistance remains a serious clinical problem. One possible reason for this problem could be alterations in signalling pathways that are downstream or parallel to ErbB-2 upon trastuzumab or lapatinib treatment. We showed that ErbB-2 inhibition activates Notch-1 that results in a compensatory increase in Notch-1-mediated proliferation (Osipo et al, 2008
). Moreover, high ErbB-2 expression in tumour-initiating cells of ErbB-2-positive breast cancer cell lines coincides with high Notch-1 expression and activity (Magnifico et al, 2009
). Interestingly, increased ErbB-2 expression in tumour-initiating cells was shown to be Notch-1 dependent (Magnifico et al, 2009
). Trastuzumab treatment was able to effectively target tumour-initiating cells of ErbB-2-positive breast cancer cell lines (Magnifico et al, 2009
). However, Notch-1 has been implicated in not only the self-renewal of these tumour-initiating cells (Magnifico et al, 2009
; Mine et al, 2009
; Harrison et al, 2010
), but also in trastuzumab resistance (Osipo et al, 2008
; Huober et al, 2010
). To circumvent these problems, we designed and evaluated for the first time a combination therapeutic strategy that can prevent and/or reverse trastuzumab resistance in vivo.
Our data provide, to our knowledge, the first preclinical proof of concept in mice for future clinical trials of combination regimens including trastuzumab and a Notch inhibitor, MRK-003 GSI, for the prevention of tumour recurrence in ErbB-2-positive breast cancer.
Notch-1, a breast oncogene, is a modulator of cell-fate decisions (Politi et al, 2004
). Overexpression of constitutively active forms of Notch-1, Notch-3, and Notch-4 develop spontaneous murine mammary tumours in vivo
(Callahan and Raafat, 2001
). Notch-4 has been shown to be critical for the survival of tumour-initiating cells (Magnifico et al, 2009
; Mine et al, 2009
; Harrison et al, 2010
). In addition, Notch-1 has been recently suggested as a novel marker of trastuzumab resistance from human breast cancer tissue (Huober et al, 2010
). We have identified Notch-1 as a novel target in trastuzumab-resistant breast cancer in vitro
(Osipo et al, 2008
). Our findings, along with evidence from the literature, indicate that Notch could be an important target in trastuzumab-resistant, ErbB-2-positive breast cancer. One class of compounds that are being used to inhibit the Notch pathway are GSIs that are currently in clinical trials for the treatment of breast cancer and other solid tumours (Pannuti et al, 2010
). Recently, it was reported that MRK-003 GSI treatment of Balb/c-neuT female mice reduced tumour onset, tumour burden, and AKT1/mTOR activities associated with ErbB-2-positive, murine breast tumours (Efferson et al, 2010
Our results demonstrate that although trastuzumab treatment caused virtually complete tumour regression (), which mimics what is observed in the clinic, it could not prevent tumour recurrence in 40% of the mice (, and ). However, adding a GSI to trastuzumab treatment either completely abolished (MRK-003) or significantly reduced tumour recurrence (, and ). Interestingly, recurrent tumours post-trastuzumab plus LY 411
575 GSI treatment showed a significant decrease in Notch transcriptional activity compared with trastuzumab treatment alone, suggesting that Notch signalling could be responsible for ErbB-2-positive breast tumour recurrence post-trastuzumab treatment. Consistent with the literature, our data could also suggest that Notch plays a critical role in the survival of tumour-initiating cells as demonstrated by the lack of tumour recurrence when Notch was inhibited in combination with ErbB-2 inhibition. Although the data presented here suggest a role for Notch based on Notch target genes modulated by a GSI, the effects on tumour recurrence could alternatively be the result of a γ
-secretase-mediated effect as it is known that a GSI specifically inhibits γ
Treatment with the combination of trastuzumab plus MRK-003 GSI simultaneously decreased proliferation and induced tumour cell death (). These antitumour effects of the combination therapy may be because of near-complete blockade of two critical signalling pathways downstream of ErbB-2: ERK1/2 and AKT1 (). Thus, a combination of trastuzumab plus MRK-003 GSI could benefit those women with recurrent, or possibly resistant, ErbB-2-positive breast cancer ultimately to reduce or eliminate disease progression and deaths by simultaneously inactivating two critical prosurvival and antiapoptotic pathways such as ERK1/2 and AKT1. The exact mechanism by which the combination treatment inhibits ERK1/2 and AKT1 activities is currently under investigation.
Lapatinib is a very potent inhibitor of ErbB-2 activity in vitro; however, lapatinib alone reduced tumour growth by only 40% in our BT474 model of ErbB-2-positive breast cancer (). A combination of lapatinib plus MRK-003 GSI showed significant reduction in the tumour growth (). This is likely because of inhibition of ERK1/2 and AKT1 activities () that resulted in increased apoptosis and decreased proliferation (). However, the onset of diarrhoea-associated toxicity with lapatinib or lapatinib plus GSI treatment at week 13 caused the study to end prematurely and, therefore, complete tumour regression was not reached.
Our results also showed that a GSI could partially restore sensitivity to trastuzumab in resistant tumours (). The mechanism by which a GSI only partially reverses trastuzumab resistance in vivo
is being actively studied. It is possible that targeting all four Notch receptors with a pan-Notch inhibitor such as a GSI might not effectively target Notch-1, which we have shown to be necessary for trastuzumab resistance in vitro
(Osipo et al, 2008
). A more specific Notch-1 or possibly other Notch signalling pathway inhibitors could prove to be more effective and potent. We are currently investigating which components of Notch signalling should be targeted in trastuzumab- or lapatinib-resistant tumours to more effectively induce tumour regression. Therapeutic targeting of Notch receptors using antibodies could prove to be a more potent and specific mode of inhibition (Wu et al, 2010
) and has not been fully investigated in breast cancer. Furthermore, the Notch pathway is complicated in tumours because of multiple modes of action. For example, it is known that Delta-like 4 on endothelial cells engages and activates the Notch-1 receptor on cancer epithelial cells to promote angiogenesis (Yan et al, 2010
). In addition, Notch signalling has been recently implicated to play a role in survival and differentiation of tumour stroma (Orr et al, 2009
). The level of complexity for the role of Notch signalling in the tumour microenvironment requires a thorough investigation of the Notch pathway in breast cancer, and most notably in anti-ErbB-2-targeted drug resistance, with the goal of identifying novel and specific targets to treat or reverse resistance.
We investigated the therapeutic benefits of two distinct inhibitors to block ErbB-2 (trastuzumab or lapatinib) and Notch (MRK-003 or LY 411
575 GSI) pathways. Although lapatinib is a more potent inhibitor of ErbB-2 activity in vitro
compared with trastuzumab, trastuzumab is more efficient in inhibiting BT474 tumour growth. As the mice used in this study are immunodeficient, this is not likely to be an immune-mediated effect (e.g., ADCC). Lapatinib plus MRK-003 GSI were not as effective as trastuzumab plus MRK-003 GSI in inhibiting tumour growth of BT474 xenografts. A combination of MRK-003 GSI plus trastuzumab was sufficient to prevent tumour recurrence, whereas a combination of LY 411
575 plus trastuzumab was only able to reduce tumour recurrence. The observed differences in efficacy between MRK-003 and LY 411
575 GSI was also observed by recent data from Grudzien et al (2010
), where the effect of MRK-003 was irreversible, leading to complete elimination of tumour-initiating cells in vitro
, whereas the effect of LY 411
575 was reversible after drug washout. These results, in addition to data from the current paper, suggest that the best combination strategy for a future clinical trial to prevent tumour recurrence of ErbB-2-positive breast cancer could be trastuzumab plus MRK-003 GSI.
In conclusion, our findings suggest for the first time that the benefit of using a combination of trastuzumab plus a GSI is prevention of ErbB-2-positive breast tumour recurrence. Because Notch is a breast oncogene that is critical for survival and proliferation of breast cancer cells, our findings strongly suggest that combined treatment with a Notch inhibitor should be an effective therapeutic strategy to prevent tumour recurrence and possibly disease progression and death in ErbB-2-positive breast cancer. Our data also suggest that a combination of trastuzumab plus MRK-003 GSI could benefit women with recurrent, or possibly resistant, ErbB-2-positive breast cancer to prevent disease progression.